Alexey Berezka
I build AI agents and automation systems that actually get deployed — and research how organizations across industries should adopt them. Engineering background at ZF means I know what works on the ground, not just in a slide deck.
I build AI systems. Then I help organizations deploy them.
AI agents, automation pipelines, multi-model workflows — I ship these things in production and run them daily. That hands-on practice informs everything else I do.
At KIT HECTOR School I'm finishing an M.Sc. researching how organizations across industries should evaluate and adopt AI — structured decision frameworks, not vendor hype. Backed by expert interviews, quantitative analysis, and a decade of engineering reality.
10 years of embedded systems, data analytics, and now AI — in automotive, aviation, finance. I understand the full stack: from infrastructure constraints to boardroom decisions.
From factory floor to AI deployment — across industries
Karlsruhe Institute of Technology
Researching how organizations should evaluate and adopt AI — structured decision framework built on expert interviews, quantitative analysis, and multi-criteria modeling. The goal: replace gut-feel AI decisions with a repeatable methodology that works across industries.
ZF Pilsen
Shipped embedded software for production ECUs in a regulated, safety-critical environment. Cut validation cycles 25% by automating HiL testing pipelines. This is where I learned what "deployed in the real world" actually means — and why it matters when advising on AI adoption.
Typhoon Research & Production Association
Designed measurement systems for aviation weather monitoring. Developed sensor-based data acquisition in a regulated environment requiring documentation rigor.
Stream
Built predictive models for customer segmentation and automated reporting pipelines — eliminating 40% of manual analyst workload. First hands-on experience turning data into decisions that the business actually acted on.
AI systems I've built and run in production
The Tree AI
Multi-agent AI system running 24/7 on my own server — voice in, voice out, smart home control, adaptive personas. Not a demo. A production AI agent I use every day.
Investment Telegram Bots
Fully autonomous AI pipeline: collects market data, runs analysis via LLM, and publishes insights to Telegram channels on schedule — zero manual intervention. A working example of AI automation delivering consistent output at scale.
Self-Hosted Cloud
Personal infrastructure on Hetzner VPS: workflow automation (n8n), DNS ad-blocking, reverse proxy — all in Docker.
3D Printing
Designing and producing 3D-printed products — refrigerator magnets of local landmarks, custom functional prints, and STL models for download.
Let's put AI to work in your organization
If you're looking for someone who builds AI agents, understands how to get them deployed, and can help your organization make the right adoption decisions — I'm available for full-time roles in Germany.